- New Method Uses Data-Driven Models to Inform COVID-19 Policymaking
This collaboration in Oncological Data and Computational Sciences will create an environment for team science through funding of trans-institutional projects in computation and oncology. An executive committee of experts from each institution will oversee the collaboration, which will have shared infrastructure and will co-recruit faculty.
Its ongoing success will depend on the development of educational programs designed to train a new generation of scientists in data and computational science and oncology.
“We have set out a vision to create a world-leading partnership to accelerate progress against cancer by combining MD Anderson’s substantial effort to digitally-enable its mission with the deep computational modeling capabilities of the teams at the Oden Institute and TACC,” said Karen Willcox, PhD, director of the Oden Institute.
“Together, the teams will represent one of the largest ecosystems of cancer care and research, expertise in computational modeling, and high-performance computing in the world.”
The initiative builds on ongoing collaborations between the Oden Institute’s Center for Computational Oncology and MD Anderson’s Department of Imaging Physics.
“Integrating oncological data with mechanism-based modeling is still a rare approach to cancer research, but these are really hard problems we’re attempting to address so new approaches are essential,” said Dan Stanzione, PhD, executive director at TACC.
“Since this initiative brings together the best minds, the best tech, and a wealth of high-quality data, we are uniquely positioned to lead in the use of computational models to understand and defeat cancer.”
A request for applications from joint MD Anderson and Oden Institute/TACC teams interested in pursuing research in data-driven computational modeling in oncology was released with projects and teams to be selected and funded by September 1, 2020.
Partnerships like these are becoming increasingly common across the healthcare industry. As technology like computational modeling and AI start to play a larger role in care delivery, institutions are collaborating to leverage data and expertise they may not otherwise have.
In May 2020, the Perelman School of Medicine at the University of Pennsylvania (Penn Medicine), Intel, and research institutions from around the world announced a partnership to use a privacy-preserving machine learning technique to identify brain tumors.
“It is widely accepted by our scientific community that machine learning training requires ample and diverse data that no single institution can hold,” said Dr. Spyridon Bakas, University of Pennsylvania.
“With this federation of 29 collaborating international healthcare and research institutions, we will be able to train state-of-the-art AI models for healthcare, using privacy-preserving machine learning technologies, including federated learning.”
By Jessica Kent